Path: utzoo!attcan!uunet!lll-winken!ames!ncar!boulder!fozzard
From: fozzard@boulder.Colorado.EDU (Richard Fozzard)
Newsgroups: comp.ai.neural-nets
Subject: Re: Data Compression
Summary: Mysterious receptive field learning
Keywords: Principal Components Analysis
Message-ID: <7537@boulder.Colorado.EDU>
Date: 18 Mar 89 15:56:17 GMT
References: <10199@nsc.nsc.com>
Sender: news@boulder.Colorado.EDU
Reply-To: fozzard@boulder.Colorado.EDU (Richard Fozzard)
Organization: University of Colorado, Boulder
Lines: 35

In article <10199@nsc.nsc.com> andrew@nsc.nsc.com (andrew) writes:

with regard to:
3b) Terence D. Sanger, "An Optimality Principle for Unsupervised Learning"

>   (I find it mysterious that random noise training produces orientation-
>   selective receptive fields spontaneously; what's the connection between
>   eigenvectors of an input autocorrelation and straight lines?
>   Not only are these fields similar to those found in retinal cells of
>   cat and monkey, but, as one goes down the list in order of decreasing
>   eigenvalue, resemble somewhat eigenstates of wave-functions of atoms
>   from quantum mechanics - perhaps a coincidental isomorphism!).
>
Well, I dont have the solution to the mystery, but Lehky and Sejnowski
report a similar learning of line segment receptive fields under equally
unexpected circumstances - learning surface curvature from shading.  This
work used standard back-prop instead of the Hebb rule, though.

Here's the reference:

"Neural Network Model for the Cortical Representation
of Surface Curvature from Images of Shaded Surfaces"
Sidney R. Lehky and Terrence Sejnowski
Department of Biophysics
Johns Hopkins University
In: Lund, J.S. (Ed.) Sensory Processing, Oxford (1988)


PS: If you talked to a certain Dr. Mandelbrot, he would insist that your
"coincidental isomorphism" was hardly that - remember fractals?


Richard Fozzard
University of Colorado
fozzard@boulder.colorado.edu